ABSTRACT Background: Older Veterans with advanced chronic kidney disease (CKD) face complex decisions to initiate or forgo dialysis in the context of uncertainty about their future health and physical function. Making these decisions is complicated by the course of advanced CKD which is characterized by frequent health events that further worsen function. Decisions support tools are needed that are specific to the clinical course of advanced CKD and predict outcomes that matter most to these patients, such as physical function. Characterizing how patients ?bounce back? from health events, such as illnesses or injuries that result in emergency department (ED) visits or hospitalizations may be key to predicting future functional status. This approach draws from the novel geriatric concept of physical resilience, defined as one?s ability to resist or recover from functional decline following a ?health stressor.? Objectives: To help older Veterans make informed decisions about kidney disease treatment by better characterizing physical resilience and identifying patient factors associated with physical resilience to develop a prediction tool for physical resilience in advanced CKD. This addresses the HSR&D priority of Patient-Centered Care domain. To do this, we propose Physical REsilience Prediction in Advanced REnal Disease (PREPARED), a prospective cohort study of older Veterans with advanced CKD with the following Aims: 1. To characterize physical function trajectories before and after an acute health stressor in order to define physical resilience among older Veterans with advanced CKD. 2. To identify associations between patient characteristics and physical resilience trajectory and potential candidate variables for prediction model development. 3. To develop a prediction tool for physical resilience (where this quantity has been defined in Aim 1). 4. To determine the association of physical resilience with short-term mortality. Methods: We will conduct a longitudinal cohort study of 800 Veterans ? 70 years old, with an estimated glomerular filtration rate (eGFR) < 30 ml/min/1.73 m2 (excluding dialysis or transplant), and 90-day probability of hospitalization ? 50% (based on the Care Assessment Needs [CAN] score). Telephone assessments will include brief validated measures of function every 8 weeks, and within 14 days following a stressor for up to 6 calls. In Aim 1, we will characterize physical resilience, first by identifying latent classes of physical resilience trajectories using general growth mixture modeling. Next, among the subset from the physical resilience latent trajectory class we will fit a piecewise linear mixed effects model to quantify resilience. In Aim 2, we will determine how the physical function trajectory is moderated by person-level health and psychosocial factors and organ system-level physiologic factors. This information will be used to identify potential candidate variables for our prediction model in Aim 3. The purpose of Aim 4 is to determine the prognostic importance of physical resilience by examining the relationship between experiencing a stressor and physical resilience with 6-month mortality. Impact: The proposed study addresses the most pressing clinical dilemma in this complex condition that disproportionately affects older Veterans. Data on physical resilience from the proposed study will be used to develop a practical tool to address a vital question that CKD patients, their families, and providers face when making treatment decisions. Limiting uncertainty about future health by predicting resilience will support individualized and patient-centered decision-making for kidney disease. Next steps: We will develop a clinical trial to test the use of our physical resilience prediction tool and work with our local and national operations partners (Durham VA Renal Service, Office of Geriatrics and Extended Care, Renal Field Advisory Committee) to implement physical resilience assessment into care for these patients.